LC_ALL=C.UTF-8 removes a spurious test failure at run-builtins line 365.
And for some tests failing due to missing locale it turns unreadable
characters into some incorrect but readable characters.
Also update known failures: the known ulimit -c failure has a different
line number; the removal of ja_JP.SJIS and the newly introduced
zh_TW.BIG5 tests also introduce several new failures.
Update to iana-etc-20250807.
Update to vim-9.1.1829.
Update to ncurses-6.5-20250809.
Update to Python-3.13.7 (Security Update).
Update to linux-6.16.1.
Update to iproute2-6.16.0.
Update to systemd-257.8.
Add a modification to glibc from upstream
to solve an incompatibility with valgrind.
A frequently asked question is "why not using -i.orig instead of -i for
${some_random_sed_command_in_book}" but I really expect people to learn
from this example and do it on their own if they really need...
XML::Parser.
It shows up as a warning while building XML::Parser, but the
functionality of the package doesn't seem to be impacted for our use
case at least.
The upstream kernel has dropped it because real 32-bit hardware almost
never has RAM > 4 GiB. The users running 32-bit software on 64-bit
hardware should really switch to MLFS now.
Regenerating the XML files also makes the prompt for X2APIC consistent
with the 6.15 kernel.
There are periodically users reporting that the --with-sysroot option
does not exist in binutils and gcc. Add a note telling that not all
aoptions are visible at the top level of the tree, unless
--help=recursive is passed. Take --with-sysroot as an example, so
that users have everything to find it...
I still think this is obvious but it seems some people are surprised by
"unknown host" in chroot. Let's just note it before they propose a
change for every BLFS package of which the build process requires an
Internet connection.
Update to readline-8.3.
Update to perl-5.42.0.
Update to openssl-3.5.1.
Update to ninja-1.13.1.
Update to linux-6.15.6.
Update to gettext-0.25.1.
Update to e2fsprogs-1.47.3.
Update to bash-5.3.
Update to iana-etc-20250618.
Update to vim-9.1.1497.
Update to util-linux-2.41.1.
Update to shadow-4.18.0.
Update to pkgconf-2.5.1.
Update to ninja-1.13.0.
Update to linux-6.15.4.
Update to less-679.
Update to automake-1.18.1.
The problem is that TIC_PATH is nor honored anymore in chapter
6, so that tic from the host is used to create the terminfo
database in chapter 6. The problem is that old versions of
tic create symlinks in the database, while newer versions
create hardlinks. Since we use a DESTDIR install in chapter 8
(with a recent version of ncurses, so with hardlinks), and copy
it in place with cp -a, then it seems that copying hardlinks to
symlinks. If an old version of tic has been used in chapter 6,
this copies hardlinks to symlinks, which creates symlinks
pointing to themselves (cp bug?).
Anyway, the solution is to copy the auxilliary tic built
in chapter 6 to $LFS/tools/bin (suggestion by Xi Ruoyao).
Now, there is no need to set TIC_PATH or whatever because this
tic is in the PATH.
Bug first reported by Marcin Dulak. Analysis with the help of
Bruce Dubbs and Thomas Trepl.
Fixes https://wiki.linuxfromscratch.org/lfs/ticket/5744
